1. Field of the Invention
The invention relates to molten metal-resistant compositions and, more particularly, to a method of manufacturing a molten metal-resistant composition employing ceramic fibers.
2. Description of the Prior art
In the casting of molten metals such as aluminum, copper, zinc, and the like, and alloys thereof, the metal generally is melted in a furnace and thereafter is transported through various transfer systems to a mold where solidification occurs. In a typical arrangement, molten metal is transferred from the furnace to a ladle and thence to a holding furnace. From the holding furnace the metal is dispensed to individual molds through a gating system employing pipes and troughs.
In order to transport the metal as described, it is necessary that the various ladles, pipe, troughs, and the like (transport elements) be protected from chemical and mechanical attack by the metal. It also is necessary that the transport elements be insulated so that the metal remains in a liquid state until solidification is desired. In response to the foregoing requirements, it has become conventional practice to provide protective liners for the transport elements. The liners desirably are of low density so that a proper insulating function is provided, but they also must have enough chemical and structural integrity to withstand attack by the metal. Transport elements that include a high percentage of ceramic fibers such as aluminosilicate fibers have been found to perform better than transport elements employing other constituents. Although ceramic fibers are subject to attack by molten metal, particularly aluminum, these fibers have been found to exhibit the best combination of insulating properties and structural integrity. Unfortunately, the techniques used to manufacture transport elements employing a high percentage of aluminosilicate fibers have certain limitations.
The most commonly used technique for forming transport elements from ceramic fibers is that of vacuum forming. In vacuum forming, a slurry consisting of water, ceramic fibers, and binders (usually a low temperature, organic binder and a high temperature, inorganic binder) is mixed in a tank. A metal mesh, or gauze, of a desired configuration is immersed in the slurry and a vacuum is applied to the mesh. Water is sucked through the mesh but solid constituents in the slurry are retained on the surface of the mesh. The vacuum is stopped when a required thickness has been deposited on the mesh. The now-formed part is removed and dried. After drying, the part can be fired at high temperature to burn out the low temperature, organic binder.
The vacuum forming process has a number of drawbacks. One of these drawbacks is that the process is a batch process, that is, only a limited number of mesh forms can be immersed in the slurry at a given time, and additional parts cannot be made until previous parts are completed. Additionally, the process can produce only parts having a uniform cross-section. If differing wall thicknesses are required, a separate machining operation is necessary. Also, it is not possible to produce articles having a length greater than about four feet. Another drawback of the vacuum forming process is that it is difficult to control the thickness of the resultant parts. Trial and error experimentation is required to approximate the desired thickness, and variations can occur from batch to batch depending upon such things as variations in the slurry, the strength of the vacuum, the length of time the vacuum is applied, and so forth.
Another technique for forming transport elements is that of lamination. In the lamination technique, layers of ceramic fiber "paper" are wrapped around a shape such as a mandrel. After enough layers have been applied, a desired wall thickness will be obtained. The principal drawback of the lamination process is that considerable effort is required to apply the paper layers properly. The technique is exceedingly difficult and time consuming to accomplish. It also is of limited usefulness in that articles having only certain shapes, such as cylinders, are amenable to manufacture by the process.
Yet an additional technique for manufacturing transport elements is that of molding. In the molding process, a thick fiber mixture is deposited into a mold. The mixture is tamped or shaped by hand to fill all parts of the mold. The molding technique is even more labor intensive than the lamination technique, and therefore is even less desirable.
Another approach is disclosed in U.S. Pat. No. 4,257,812. In the '812 patent, aluminosilicate fibers, kaolin clay, plasticizers, water, and lubricants are mixed together and then are extruded into a desired shape. Thereafter, the shape is dried to remove so-called mechanical water, and thereafter is fired at elevated temprature to drive off so-called chemical water and transform the kaolin clay into the metakaolin phase. Although the extrusion process disclosed in the '812 patent offers certain advantages over certain other processes such as the lamination and molding processes, it fails to address certain concerns. One of those concerns relates to the firing temperature. If the firing temperature inadvertently should exceed approximately 1800.degree. F., the metalkaolin will abruptly decompose and form alumina, mullite, and free silica, substances of no use in protecting against attack by molten metal. Even an inadvertent, temporary firing at approximately 1800.degree. F. will ruin the extruded shapes.
An additional problem relates to the density of the extruded product. In order to obtain desirable insulating properties, the extruded product should have a low density on the order of 35-45 lbs. per cubic foot. However, the density of the resultant product disclosed in the '812 patent is on the order of 95-100 lbs. per cubic foot. This is believed to be caused by a relatively small percentage of water used to produce an extrudable mixture.
Desirably, a process for manufacturing molten metal-resistant ceramic fiber compositions would be available that would yield products having a lower density than products produced by prior extrusion processes. The process also hopefully would have advantages over prior possesses such as vacuum forming, including better dimensional tolerances, the capability of producing products having longer lengths and variable wall thicknesses, and improved manufacturability.